TABLE 1.
Authors, y | Patient Population (No.) | Steroid Intervention | Study Type (Statistical Analysis) | Outcomes |
---|---|---|---|---|
The Boston Collaborative Drug Surveillance Program, 197218 | Hospitalized medical (718) | Prednisone mean dose 32.2 mg/d | Prospective (observational study) | Psychiatric reactions reported in relation to dose: ≤ 40 mg (1.3%), 41–80 mg (4.6%), > 80 mg (18.4%) |
Rundell and colleagues, 198919 | Hospitalized with psychiatric consultation (755) | Corticosteroids not otherwise defined | Retrospective review (descriptive) | Steroids prescribed in 53.8% (7 of 13) of patients with manic syndrome |
Gaudreau and colleagues, 200520 | Hospitalized cancer (261) | Dexamethasone equivalents > 15 mg | Prospective cohort (multivariate Cox regression model) | Development of delirium was associated with steroid exposure (HR 2.67; 95% CI, 1.18–6.03) |
Gaudreau and colleagues, 200721 | Hospitalized cancer (114) | Dexamethasone equivalents > 15 mg | Prospective cohort (χ2 proportions) | Odds of probability of delirium were not significant with higher doses of steroids (P = .12) |
Schreiber and colleagues, 201422 | Medical/surgical ICU (520) | Prednisone equivalents median 44 mg (IQR 13–75 mg) | Prospective cohort (first-order Markov logistic regression model) | Steroids independently associated with the transition to delirium (OR, 1.52; 95% CI, 1.05–2.21) |
Wolters and colleagues, 201523 | Medical/surgical ICU (1,112) | Prednisone equivalents median 50 mg (IQR 25–75 mg) | Prospective cohort (first-order Markov logistic regression model) | Steroids were not associated with the transition to delirium (OR, 1.08; 95%CI, 0.89–1.32) |
Matschke and colleagues, 201624 | Hospitalized, immune thrombocytopenia (26) | Prednisone 1–2 mg/kg/d vs pulsed dexamethasone 0.6 mg/kg/d | Prospective, randomized comparative (Fisher exact test of proportions) | Dexamethasone patients reported significantly more insomnia than did prednisone patients (P = .03) |
Tilouche and colleagues, 201825 | Medical ICU (206) | Corticosteroids not otherwise defined | Prospective, observational (multivariate regression analysis) | Steroid use was independent predictor of developing delirium (OR, 2.8; 95% CI, 1.06–7.28) |
Abbreviations: HR, hazard ratio; ICU, intensive care unit; IQR, interquartile range; OR, odds ratio.